Phase 1 clinical trial of irradiated autologous melanoma cells adenovirally transduced with human GM-CSF gene

Abstract

The objective of this study was to determine the safety and antitumor activity of an autologous GM-CSF-secreting melanoma cell vaccine that was engineered ex vivo with recombinant replication-incompetent adenovirus harboring a human GM-CSF gene (Adv/hGM-CSF). Melanoma samples were surgically obtained from 30 patients (15 female and 15 male, ages ranging from 23 to 87) and were processed for vaccine preparation. Due to stringent eligibility criteria, 9 out of 30 patients were enrolled in the phase 1 clinical trial (FDA IND7677). Melanoma cell lines established from surgical specimens of 9 patients were transduced with Adv/hGM-CSF (MOI of 100) and subsequently irradiated at 35 Gy. These cell lines secreted human GM-CSF in vitro at an average rate of 80-424 ng/10(6) cells/24 h. All patients were intradermally and subcutaneously injected at several sites with irradiated autologous melanoma cells (2x10(6)-1x10(7) in 300 microl saline), 2-10 times, at intervals of 4-8 weeks. None of the patients vaccinated showed any serious adverse systemic response. Three patients (nos.1, 6 and 7) demonstrated local reaction (erythema) to the vaccination. Tumor-specific CTL assays performed in the absence of K562 cells showed that the levels of CTLs in peripheral blood of 5 patients increased following vaccination, whereas those in one patient declined. Levels of CTLs assayed in the presence of K562 cells were considerably lower than those assayed in the absence of K562 cells, but were also found to increase following vaccination in the peripheral blood of 6 patients. A patient who had been vaccinated 10 times (patient 1) responded to the vaccination by apparent reduction in size of metastatic tumor in the lung. Immunohistochemical examination of the vaccination sites of patient 1, biopsied after the 3rd and 4th vaccination. showed that the vaccination sites responded with infiltration of inflammatory cells, such as T cells (CD3+, CD8+), macrophages and dendritic cells (CD83+), for a period up to about 8 days. These data suggest that repeated vaccinations with irradiated autologous GM-CSF-producing tumor cells were well tolerated by patients and led to the activation of an antitumor immune response in some patients.

Phase 1/2 trial of autologous tumor mixed with an allogeneic GVAX® vaccine in advanced-stage non-small-cell lung cancer

“…[1][2][3] Preclinical data also suggest that a threshold of vaccine-associated GM-CSF secretion is necessary to induce optimal antitumor immunity. 4 Clinical trials have been performed in a variety of tumor types, including melanoma, [5][6][7] prostate, 8 renal cell, 9 pancreatic, 10 and lung cancer. 11,12 Vaccine treatment has been consistently well tolerated and clinical activity has been demonstrated.…”

Cell encapsulation technology as a novel strategy for human anti-tumor immunotherapy

“…Owing to biosafety concerns, this genetic modification method is preferable over a retroviral infection approach because retroviral infection involves risks of development of competent recombinant retroviral particles in vivo. K562-hGM-CSF cells secreted approximately 3000 ng per 10 6 cells every 24 h, which is higher than previously reported values that range from 42 to 1403 ng per 10 6 cells every 24 h. 19,22,41,42,44,45,47 Although the GM-CSF threshold for an effective immunostimulation effect has been evaluated to be 36 ng per 10 6 cells every 24 h, 48 there are no clearly defined values reported for the site of cell injection. On the other hand, Serafini et al 49 reported that a high-dose of GM-CSF could impair the immune response through the recruitment of myeloid suppressor cells.…”

“…This immunostimulatory strategy has been widely studied in murine models and phase I clinical trials using irradiated tumor cells genetically engineered to produce GM-CSF. 10,15,19,[21][22][23][41][42][43] The inability to obtain reproducible levels of the cytokine, and the labor-intensive requirements for individualized genetic engineering, led to the development of allogeneic bystander cells for the local delivery of GM-CSF, such as the K562 human erythroleukemia cell line. [44][45][46] However, despite their undetectable expression of human leukocyte antigen class I and II, the high sensitivity of allogeneic K562 cells to natural killer cytotoxicity prevented continuous release of the cytokine for several days.…”

Necrotic death but not irradiation abolishes costimulation of T‐cell effector functions and survival by CD80‐expressing tumor cells

“…[9][10][11]14,[29][30][31][32][33] It is unknown, however, whether the maximal tumor-specific T-cell response is achieved by living tumor cells or by cell vaccines inactivated through irradiation or freeze/thaw treatment. Here, we show that living tumor cells and tumor cells inactivated by irradiation induce similar T-cell effector responses, i.e., T-cell proliferation, inhibition of AICD and tumor cell lysis.…”

“…8 Similar to preclinical murine models, human autologous tumor cells or allogeneic tumor cell lines have been employed as vaccines in clinical trials. [9][10][11][12][13][14] Such tumor cell vaccines appear to be interesting in neoadjuvant settings, i.e., where the main tumor load has been removed and putatively only minimal residual disease remains. For fear of inducing new tumor burden, tumor cell vaccines in humans are generally inactivated before inoculation.…”